Skip to content

Commit 6446db5

Browse files
puyanguvicpuyanguvic
committed
update the docs
1 parent 046875d commit 6446db5

87 files changed

Lines changed: 606 additions & 131 deletions

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

‎README.md‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,4 @@
1-
# Agris: Phish Email Detection Agent
1+
# Argis: Phish Email Detection Agent
22

33
A minimal, deterministic phishing detection agent built around evidence-first design. The system collects structured evidence, quantifies risk, fuses evidence with policy rules, and only then produces a verdict. Models (if added later) are just evidence sources, never the final judge.
44

@@ -38,7 +38,7 @@ phish-agent/
3838
└── tests/ # Unit tests
3939
```
4040

41-
Protocol details: `docs/protocol_v1.md`.
41+
Protocol details: `docs/get-started/concepts/protocol/v1.md`.
4242

4343
## Quick start
4444

‎docs-site/_config.yml‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,8 @@
11
title: Argis Docs
22
description: Phish Email Detection Agent documentation.
33
theme: null
4+
plugins:
5+
- jekyll-redirect-from
46
markdown: kramdown
57
kramdown:
68
input: GFM

‎docs-site/_data/nav.yml‎

Lines changed: 8 additions & 24 deletions
Original file line numberDiff line numberDiff line change
@@ -1,26 +1,10 @@
11
- title: Overview
22
url: /
3-
- title: Introduction
4-
url: /overview/
5-
- title: Architecture
6-
url: /architecture/
7-
- title: Agent
8-
url: /agent/
9-
- title: Skills
10-
url: /skills/
11-
- title: Detection Pipeline
12-
url: /pipeline/
13-
- title: Reporting
14-
url: /reporting/
15-
- title: Policies
16-
url: /policies/
17-
- title: Evaluation
18-
url: /evaluation/
19-
- title: Deployment
20-
url: /deployment/
21-
- title: Developer Guide
22-
url: /developer/
23-
- title: Governance
24-
url: /governance/
25-
- title: Glossary
26-
url: /glossary.html
3+
- title: Get Started
4+
url: /get-started/
5+
- title: Using Argis
6+
url: /using-argis/
7+
- title: Configuration
8+
url: /configuration/
9+
- title: Releases
10+
url: /releases/

‎docs-site/_layouts/default.html‎

Lines changed: 6 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -21,9 +21,10 @@
2121
</button>
2222
<nav class="top-nav" aria-label="Primary">
2323
<a href="{{ "/" | relative_url }}">Docs</a>
24-
<a href="{{ "/architecture/" | relative_url }}">Architecture</a>
25-
<a href="{{ "/pipeline/" | relative_url }}">Pipeline</a>
26-
<a href="{{ "/reporting/" | relative_url }}">Reporting</a>
24+
<a href="{{ "/get-started/" | relative_url }}">Get Started</a>
25+
<a href="{{ "/using-argis/" | relative_url }}">Using Argis</a>
26+
<a href="{{ "/configuration/" | relative_url }}">Configuration</a>
27+
<a href="{{ "/releases/" | relative_url }}">Releases</a>
2728
</nav>
2829
</header>
2930

@@ -49,10 +50,10 @@ <h1>Phish Email Detection Agent</h1>
4950
operational guidance for secure email pipelines.
5051
</p>
5152
<div class="hero-actions">
52-
<a class="button primary" href="{{ "/overview/problem-statement.html" | relative_url }}">
53+
<a class="button primary" href="{{ "/get-started/overview/problem-statement.html" | relative_url }}">
5354
Start with the problem
5455
</a>
55-
<a class="button ghost" href="{{ "/architecture/system-overview.html" | relative_url }}">
56+
<a class="button ghost" href="{{ "/get-started/concepts/architecture/system-overview.html" | relative_url }}">
5657
See system overview
5758
</a>
5859
</div>

‎docs/configuration/agents-md.md‎

Lines changed: 14 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,14 @@
1+
---
2+
layout: default
3+
title: AGENTS.md
4+
---
5+
6+
# `AGENTS.md`
7+
8+
`AGENTS.md` files are local, human-authored instructions for contributors and automation.
9+
10+
This repo uses them to enforce safe handling of email content (treat headers/bodies as sensitive), prefer offline/deterministic tooling, and keep protocol boundaries stable (`protocol/`).
11+
12+
## Scoping rule (important)
13+
14+
An `AGENTS.md` applies to the directory tree rooted at its folder. Nested `AGENTS.md` files override parent scope rules.

docs/deployment/configuration.md renamed to docs/configuration/config-file.md

Lines changed: 15 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,9 +1,11 @@
11
---
22
layout: default
3-
title: Configuration
3+
title: Config File
4+
redirect_from:
5+
- /deployment/configuration.html
46
---
57

6-
# Configuration: YAML / ENV
8+
# Config file: YAML / ENV
79

810
The goal is to tune routing, weights, and thresholds without code changes, and keep versions traceable.
911

@@ -20,6 +22,8 @@ Provider/connector examples:
2022
- `configs/connectors/gmail.yaml`
2123
- `configs/connectors/imap.yaml`
2224

25+
Note: provider/connector configs exist, but the Gmail/IMAP connectors and the Ollama provider are placeholders in this build (they raise `NotImplementedError` when invoked).
26+
2327
Key blocks (`engine/config.py`):
2428

2529
- `router`: `t_fast/t_deep`, tools per profile, budgets, fallback
@@ -28,6 +32,15 @@ Key blocks (`engine/config.py`):
2832
- `allowlist_domains`: org allowlist domains
2933
- `contextual_escalation`: collaboration/OAuth escalation triggers
3034

35+
## 1.1) Resolution order (engine)
36+
37+
When using the protocol-driven engine (`engine/argis.py`), config resolution is:
38+
39+
1. `SessionConfig.profile` → `configs/profiles/{profile}.yaml` (if exists)
40+
2. `SessionConfig.config_path` (if provided)
41+
3. `configs/app.yaml` selectors (`profile_path` or `profile`)
42+
4. Fallbacks: `configs/profiles/balanced.yaml` → `configs/default.yaml`
43+
3144
## 2) ENV (recommended for extensions)
3245

3346
Current core does not depend on env vars. If online tools/secrets are added:

‎docs/configuration/custom-prompts.md‎

Lines changed: 19 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,19 @@
1+
---
2+
layout: default
3+
title: Custom Prompts
4+
---
5+
6+
# Custom prompts
7+
8+
This build is **deterministic by default** and does not ship a prompt-tuning surface yet.
9+
10+
If/when provider-backed LLM steps are introduced, this page should document:
11+
12+
- where prompts live (config vs. code)
13+
- how prompts are versioned and reviewed
14+
- how prompt outputs are recorded/redacted
15+
16+
## Where prompt code would plug in
17+
18+
- Provider interface: `providers/model/base.py` (`ModelProvider.generate()`).
19+
- Orchestrator: keep tool outputs structured and route them into `EvidenceStore` fields before scoring/rules.

docs/developer/adding-skills.md renamed to docs/configuration/extending/adding-skills.md

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,8 @@
11
---
22
layout: default
33
title: Adding Skills
4+
redirect_from:
5+
- /developer/adding-skills.html
46
---
57

68
# Adding a skill

docs/developer/adding-tools.md renamed to docs/configuration/extending/adding-tools.md

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,8 @@
11
---
22
layout: default
33
title: Adding Tools
4+
redirect_from:
5+
- /developer/adding-tools.html
46
---
57

68
# Adding tools

docs/developer/code-structure.md renamed to docs/configuration/extending/code-structure.md

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,8 @@
11
---
22
layout: default
33
title: Code Structure
4+
redirect_from:
5+
- /developer/code-structure.html
46
---
57

68
# Code structure

0 commit comments

Comments
 (0)